(This blog post is adapted from a presentation made by Natalie Enright Jerger at the ISCA 2019 business meeting.)

What is CAL?

IEEE Computer Architecture Letters (CAL) was established 17 years ago to provide the computer architecture community with a rapid dissemination venue for highly novel, highly impactful ideas.  CAL is intentionally different from both the conferences and journals in our field. With only a 4-page budget, CAL papers focus more on ideas and less on evaluations. As such, most CAL papers can and should be later published as full-length conference papers with more detailed descriptions and in-depth evaluations. CAL provides a first decision on submitted papers within 30 days.  This is considerably shorter than our current conference review cycles. In the R2 survey that was undertaken a few years ago, 45% of the community felt that the review process was too long. CAL papers appear online within 1 week of acceptance—ready to download and be read and cited by the community. Compare this to the typical 3-month lag between conference acceptance and the publication of the conference proceedings.  In summary, CAL is a great venue for architects because it:

• Focuses on high-impact, high-novelty ideas in a short, easy-to-read format,
• Provides a quick review turnaround allowing you to timestamp your idea in an archival format,
• Rapidly posts accepted papers online so that your work can quickly be cited and built upon, and
• Is one of the few available journals that focuses on computer architecture; some institutions continue to prize journals highly and this provides architects at those institutions with an excellent journal for dissemination.

Current situation

IEEE Computer Society (CS) periodically performs a review of their publications portfolio and identifies “under-performing” publications that should be discontinued.  As communities grow and shrink and topics become more or less relevant, IEEE CS sensibly adjusts its portfolio to effectively use its limited resources and serve the growing and dynamic IEEE CS community.  

While CAL has the support of the computer architecture community, it has been identified as under-performing and is slated for potential sunset in the near future. To prevent the sunsetting of this valuable forum for the computer architecture community, we as a community need to work quickly and put some effort into improving CAL along a number of dimensions. 

What is the CAL leadership doing?

The CAL editorial board, with valuable input from leaders within the architecture community (past EICs of CAL, chairs of IEEE TCCA and TCuarch, and experienced leaders within IEEE CS), has been brainstorming how to improve CAL.  Some of these improvements can be instituted by the editorial board (now), but others will require your help (next).

We have two primary goals.  1) Increase impact: visibility, downloads, readership, and citations and 2) Decrease submission-to-publication (“sub-to-pub”) latency.   

• To improve the impact of CAL papers, we must increase both readership and paper quality.   A paper must be of high quality and it must be read for it to have impact. Authors (you!) control quality, but we can increase visibility and readership. Improved visibility for your papers will lead to increased downloads, readership, and citation counts.  Increased visibility will also make CAL a more attractive venue for you to submit your high quality work.
  • We will improve the publicity for published papers via emails to subscriber lists and posts on social media.    
  • We will attract more attention to the research published at CAL through lightning talks associated with papers.  As YouTube lightning talks of conference presentations have become more mainstream, we feel that CAL papers could also benefit from brief videos that highlight the key ideas and novelty. 
  • We will better publicize the Best of CAL session at HPCA.
  • We will work with a new marketing/publicity staff member at IEEE CS on a plan to further improve CAL’s visibility.
  • We will try to attract more strong papers through partnerships with computer architecture workshops. 
• To improve our sub-to-pub latency, we must improve the review process and reviewer latency (you, again!).  
  • We will replace the decision “Major Revision” with the decision “Revise and Resubmit.” This alternative still provides continuity of editor and reviewers without including the revision time in the sub-to-pub time.   Most Letters do not feature a “Major Revision”, nor did CAL in its original incarnation. Removing the “Major Revision” option will help CAL deliver on its goal of rapid publication. 
  • We will perform more frequent analysis of where review time is going and let go of Associate Editors who are not staying on top of their papers.

What can you do?

Despite these initiatives, CAL needs your help to become a more successful publication.  We ask you to:

• Consider more frequently submitting your work to CAL.  Increasing the number of submissions will clearly demonstrate the value of CAL to the community and to IEEE.  
• Respond quickly to review requests and complete your reviews in a timely fashion.  
• Read CAL papers via downloads from IEEExplore.  
• Cite CAL papers alongside their subsequent conference versions. 

A straw poll at the ISCA 2019 business meeting showed an overwhelming number present wanted to save CAL.  Raising your hand is easy, but if you truly want to save CAL, the best way to help is to perform the actions above.

What do you think?

As we embark on this process to save CAL, it is important to assess the community’s attitude toward CAL.   Research dissemination has changed rapidly over the last decade. Notably, the rise of arXiv as a means to timestamp ideas and start collecting citations might lead one to question whether the community still needs CAL.  We believe the answer is yes; CAL provides a rigorous peer review and is a highly competitive venue. A CAL paper carries greater weight than an arXiv paper; the latter is more akin to a technical report than a full-fledged publication.  However, we understand that some might disagree, and we welcome your feedback on whether you think CAL is still important today.  

Please comment on this blog or write to the EIC (sorin@ee.duke.edu) with your thoughts.   We are always open to feedback and suggestions, and we gladly welcome volunteers to help us in our efforts to save CAL.  

About the authors: Daniel J. Sorin is a Professor of Electrical and Computer Engineering at Duke University and currently serves as the Editor-in-Chief for CAL. Natalie Enright Jerger is the Percy Edward Hart Professor and Canada Research Chair in Computer Architecture in the ECE department at the University of Toronto. She currently serves as the Associate Editor-in-Chief for CAL. 

Disclaimer: These posts are written by individual contributors to share their thoughts on the Computer Architecture Today blog for the benefit of the community. Any views or opinions represented in this blog are personal, belong solely to the blog author and do not represent those of ACM SIGARCH or its parent organization, ACM.